ChunhuanLin/deform_conv_pytorch

Issues

• paper benchmark？

  #21 opened 3 years ago by b762927
  0

• Visualize offsets

  #6 opened 7 years ago by Youngkl0726
  4

• What is the offset value along the input for forward() in DeformConv2D()

  #20 opened 4 years ago by shahzad-ali
  0

• Why did you permute p = p.contiguous().permute(0, 2, 3, 1)?

  #19 opened 5 years ago by kanopimurah
  0

• ValueError: cannot reshape array of size 18 into shape (1,128,1,1)

  #14 opened 6 years ago by Esaada
  3

• RuntimeError: Invalid index in gather at /pytorch/aten/src/TH/generic/THTensorEvenMoreMath.cpp:657

  #18 opened 5 years ago by Meimeiainaonao
  0

• meaning of q_lt, g_lt and x_q_lt

  #17 opened 5 years ago by Damax18
  0

• report a bug

  #16 opened 5 years ago by nbl97
  0

• why use mask in the zero padding area？

  #15 opened 6 years ago by hjy1312
  1

• Different interpretations of def_conv

  #10 opened 6 years ago by bkvie
  4

• A problem about dilation and stride settings

  #13 opened 6 years ago by rxqy
  1

• padded offset

  #11 opened 6 years ago by bkvie
  1

• ValueError: cannot reshape array of size 18 into shape (1,64,1,1)

  #12 opened 6 years ago by Meimeiainaonao
  3

• Stride setting error in DeformConv2D?

  #9 opened 6 years ago by smilewsw
  1

• Only 1 deformable convolution

  #4 opened 6 years ago by bkvie
  3

• bias for offset and deformable conv

  #5 opened 6 years ago by zhujiagang
  2

• Some uncertainty about the demo

  #3 opened 6 years ago by fatterZhang
  5

• Deformable ROI pooling layer

  #8 opened 6 years ago by Karthik-Suresh93
  1

• typo

  #7 opened 7 years ago by leao1995
  1

• change train loader in demo.py

  #2 opened 7 years ago by bkvie
  1

• Cuda memory error when perform evalution

  #1 opened 7 years ago by John1231983
  7